tubular tower with thin wall

[rcat]

I'am trying to model a long cylinder (a tubular tower with thin wall -- d/t>120) subject to a uniform distributed load in all the length. Given the importance of the imperfections in the analysis, its possible to take them in account?

Those values aren't purely empirical?

Regards,

rcat

 

[MSDw]

They are probablistic and come from statistical studies, and usually taken as tolerances in any design. you can check for vertical tolerances in any building code.

 

[rb1957]

could you make your surface wavy, rather than straight (ie cyclindrical) ... and define a waviness criteria on the manufacture of the tube. then you could look at the effect of different wavelengths, and maybe even phasing the crests (so the surface has almost a spiral look to it).

 

[crisb]

It should be possible to do an eigenvalue buckling run and use the results as initial imperfections for a nonlinear run (using an appropriate factor to give a maximum displacement to a relevant code).

 

[JStephen]

Those imperfections would be important primarily if buckling is the concern? Is that the case?

 

[GregLocock]

The way we'd tackle it is to model it several times with different imperfection sizes and build up some knowledge of how robust the model is against typical real imperfections.

For example - are loads truly uniform? Is the material uniform? Are the boundary conditions important?

Until you hav shown that you understand these things I'd say your model is a pretty picture.





Cheers

Greg Locock

Please see FAQ731-376 for tips on how to make the best use of Eng-Tips.

 

[corus]

A DIN standard gives factors to allow for imperfections when assessing a structure for buckling.

corus

 

[rcat]

Thanks for all for your replies (and sorry for the bad English).

The European codes don’t contain any relevant information for this specific case and lots of investigation has been made about it. Look the article:

http://dol.dl.uni-leipzig.de/servle...C9F8C66CA6CDE2ED0C3CCFB.tomcat_01?hosts=local

Some more studies in real towers will be sent to me to compare with numerical models. There is lots of work undone.

Iam an assiduous reader of the group but my poor english disable my participation.

rcat

 

[mtnengr]

Rcat: Don't worry, for many of us our German is worse than your English.

There have been several attempts to examine imperfection geometry effects on critical buckling loads. Look at the references at my web site

http://www.volcano.net/~d.citerley.

You will find computer programs that will help you in the evaluation and papers on the subject (WRC Bull 313). If you need any of the references, I can forward you a PDF file of a given paper.

 

[rcat]

Thanks, Mtnengr.

Iam interested in the reading of WRC Bulletin 313.
If possible, please forward to: rtproj@hotmail.com